OBJECTIVE: To compare the accuracy of three-dimensional computed tomography (3D-CT) and panoramic radiography in the evaluation of mandibular hypoplasia in patients with hemifacial microsomia (HFM). DESIGN: Retrospective study of imaging data. Setting : Images selected from the archives of the University of Tokyo Hospital. SUBJECTS: Twenty patients with unilateral HFM who had undergone both panoramic radiography and 3D-CT in the same period. METHOD: Mandibular deformities were classified according to the Pruzansky classification; eight patients had Grade I deformity and 12 patients had Grade II deformity. Ramus heights were measured on both panoramic radiographs and 3D-CT. MAIN OUTCOME MEASURES: Magnification in panoramic radiographs and extent of mandibular asymmetry as estimated by the affected/unaffected side ratio based on two methods were examined. The Pearson product-moment correlation coefficient was used to estimate correlations between parameters. RESULTS: The magnification of ramus heights on panoramic radiographs showed large variations in Grade II patients. The affected/unaffected side ratio estimated by the two methods showed a strong correlation in Grade I patients (correlation coefficient 0.99; p < .0001). Conversely, a weak correlation was seen in Grade II patients (correlation coefficient 0.77; p = .0036), and affected/unaffected side ratios from panoramic radiographs were both over- and underestimated. CONCLUSIONS: The accuracy of evaluation using panoramic radiography was fairly reliable in Grade I patients. Conversely, accuracy was poor in Grade II patients, and evaluation using 3D-CT seems preferable. The combination of two methods with careful consideration is recommended for clinical applications.
OBJECTIVE: To compare the accuracy of three-dimensional computed tomography (3D-CT) and panoramic radiography in the evaluation of mandibular hypoplasia in patients with hemifacial microsomia (HFM). DESIGN: Retrospective study of imaging data. Setting : Images selected from the archives of the University of Tokyo Hospital. SUBJECTS: Twenty patients with unilateral HFM who had undergone both panoramic radiography and 3D-CT in the same period. METHOD: Mandibular deformities were classified according to the Pruzansky classification; eight patients had Grade I deformity and 12 patients had Grade II deformity. Ramus heights were measured on both panoramic radiographs and 3D-CT. MAIN OUTCOME MEASURES: Magnification in panoramic radiographs and extent of mandibular asymmetry as estimated by the affected/unaffected side ratio based on two methods were examined. The Pearson product-moment correlation coefficient was used to estimate correlations between parameters. RESULTS: The magnification of ramus heights on panoramic radiographs showed large variations in Grade II patients. The affected/unaffected side ratio estimated by the two methods showed a strong correlation in Grade I patients (correlation coefficient 0.99; p < .0001). Conversely, a weak correlation was seen in Grade II patients (correlation coefficient 0.77; p = .0036), and affected/unaffected side ratios from panoramic radiographs were both over- and underestimated. CONCLUSIONS: The accuracy of evaluation using panoramic radiography was fairly reliable in Grade I patients. Conversely, accuracy was poor in Grade II patients, and evaluation using 3D-CT seems preferable. The combination of two methods with careful consideration is recommended for clinical applications.